How Long People Have Lived in Australia, and How Archaeologists Know

The most widely cited archaeological figure is around 65,000 years, from excavations at Madjedbebe in northern Australia published in Nature in 2017. The authors of that paper present it as a minimum age rather than a final answer, and open by stating that the time of arrival of people in Australia is an unresolved question. Aboriginal and Torres Strait Islander peoples have lived on the continent continuously ever since, and that continuity — unbroken occupation, language, law and connection to Country through to the present — is not in dispute in any part of the literature.

What the excavated sites show

Madjedbebe. The 2017 report describes artefacts found in primary depositional context, concentrated in three dense bands, with the integrity of the deposit demonstrated by refitting broken artefacts back together and by optical dating and other analyses of the sediments. Occupation began around 65,000 years ago, associated with a distinctive stone tool assemblage: grinding stones, ground ochres, reflective additives and ground-edge hatchet heads. The authors frame the result as setting a new minimum age for human arrival in Australia and, by extension, for the dispersal of modern humans out of Africa.

Lake Mungo. In the Willandra Lakes region of New South Wales, the Mungo remains include what the published literature describes as the world's oldest ritual ochre burial and the first recorded cremation. Their dating history is a useful lesson on its own. Before the 2003 study, the burial known as Mungo III had been variously estimated at 30,000 years, at 42,000–45,000 years, and at 62,000 ± 6,000 years; radiocarbon estimates had placed the Mungo I cremation near 20,000–26,000 years. A new series of 25 optical ages published in Nature concluded that both burials occurred at 40,000 ± 2,000 years ago and that people were present at Lake Mungo by 50,000–46,000 years ago. The paper records that the work was carried out with the support of the three traditional tribal groups of the Willandra Lakes World Heritage Area.

That is a spread of more than thirty thousand years across successive published estimates for one site. Anyone quoting a single number for Australian occupation should know that this is what the underlying literature looks like.

Why optical dating changed the picture

Radiocarbon dates organic material. Optically stimulated luminescence dates the sediment itself.

Mineral grains such as quartz accumulate a trapped-charge signal from low-level background radiation while they are buried. Exposure to daylight drains that signal to zero. So if a grain is buried in the dark, the signal starts building again from the moment of burial, and measuring how much has accumulated gives an estimate of how long ago that grain last saw the sun.

Two features matter for archaeology. It dates layers rather than objects, so a site with no surviving bone or charcoal can still be dated. And in single-grain form it can detect mixing: if grains within one apparent layer return wildly different ages, the deposit has been disturbed, and the excavator can say so rather than averaging the problem away. The Madjedbebe chronology rests on a statistical model of single-grain optical ages for exactly this reason.

The trade-off is that it dates burial, not human activity. Connecting the two is a separate inferential step, which is why the artefact refits and stratigraphic checks in that paper carry as much weight as the ages themselves. How historians date events covers that gap between what a method measures and what a claim asserts, across methods.

Why institutional figures differ

UNESCO's World Heritage inscription for Kakadu National Park describes the reserve as having been inhabited continuously for more than 40,000 years, with cave paintings, rock carvings and archaeological sites recording the skills and way of life of the region's inhabitants. Parks Australia's own page for the same park, written in the voice of the Traditional Owners, states that Kakadu is a living cultural landscape and has been home to Aboriginal people for more than 65,000 years, and identifies the Bininj people of the park's north and the Mungguy people of its south.

Both are official. They differ because they were written decades apart, and the archaeological figure moved in between. This is a good habit to carry generally: when two authoritative sources disagree about a prehistoric date, check the vintage of each document before concluding that anyone is wrong.